MurphyML

My machine learning study project

Project setup

Using Pyenv + Anaconda

Install pyenv

brew install pyenv

Setup pyenv

Setup your ~/.zshrc or ~/.bash_profile depending on your shell. For detailed settings. Refer to following command.

pyenv init

Check available anaconda versions

pyenv install --list | grep anaconda

Install anaconda

pyenv install anaconda3-5.3.1

Using venv

Create venv

If you are using python with newer version than 3.3, you can create virtual environment by

python3 -m venv .

Activate VM

source bin/activate

Dependencies

All dependencies are described in requirements.in file.

To install dependencies, use following commands

pip3 install -r requirements.txt

Generate requirements.txt file

First, install pip-tools

python3 -m pip install pip-tools

Specify requirements in requirements.in file. Then run following command to update requirements.txt.

pip-compile

Tests

We use pytest for unit testing. Use following commands to run tests.

pytest # run all tests.
pytest test # run all tests in specific package.
pytest test/sample_test.py # run tests in specific module.
pytest test/sample_test.py -k test_sample # run specific test case.
pytest test/sample_test.py::SampleTest::test_sample # run specific test case.

ML Framework

We used Pytorch for ML research.

https://github.com/pytorch/pytorch